The mediating role of psychological distress in the association between migration experiences and stigma and coping strategies: a cross-sectional study on Ethiopian migrant returnees

Abstract

Background

Ethiopian migrants to the Middle East and South Africa are exposed to dangerous traveling and working conditions and their experiences are mostly tragic. They are unwelcomed not only by the situation in the destination but also by the community at home which is an important indicator of stigma and discrimination. However, there is lack of evidence on how psychological distress is associated with migration experiences, stigma and coping strategies. Therefore, it was aimed to determine the mediating effect of psychological distress in the association between migration experiences and stigma and discrimination and coping strategies.

Methods

A total of 739 Ethiopian migrant returnees from the Middle East and South Africa were included in the study. A cross-sectional study was conducted in five migration hotspot areas in Ethiopia, namely Addis Ababa, Dessie, Shashemene, Hossana, and Gondar. Data related to socio-demographic, economic, migration experiences, psychological distress, coping strategies, and stigma and discrimination were collected. Psychological distress was measured using the 21- item version of the Depression, Anxiety and Stress Scale and coping strategies were measured with the Brief-COPE. Structural equation modeling was employed to estimate the relationship among variables.

Results

About 395 (57.33%) of the participants reported symptoms of depression, 428 (59.86%) anxiety symptoms, and 313 (45.21%) stress symptoms. The mediated association between physical violence and coping strategy through psychological distress was 0.29 (adjusted β = 0.29, 95%CI: 0.15, 0.44). Denial of salary had positive direct (adjusted β = 1.00, 95%CI: 0.50, 1.50) and mediated (adjusted β = 1.20, 95%CI: 0.71, 1.68) associations with stigma and discrimination through psychological distress. Restricted freedom had positive mediated (adjusted β = 0.39, 95%CI: 0.13, 0.65) and total (adjusted β = 0.94, 95%CI: 0.59, 1.29) associations with stigma and discrimination through psychological distress.

Conclusion

Depression, anxiety, and stress symptoms are common among migrant returnees. Religious practice is the most commonly used coping mechanism. Psychological distress has positive mediating association with the relationships between denial of salary by employers and stigma and discrimination, physical violence and coping strategy, and restricted freedom and stigma and discrimination. Psychological interventions to deal with the psychological distress and stigma and discrimination of migrant returnees and to encourage the practice of positive coping strategies are warranted.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12888-024-06229-6.

Introduction

The magnitude of transnational migration is significantly increasing over time, and one reason for this is globalization as it facilitates the porosity of international boundaries [1, 2]. Migration that is demand-driven and carefully managed benefits both the migrants and the host country. However, its negative consequences might outweigh when there is mismatch of needs between the two parties. In such conditions, migrants are unlikely to get jobs with safer working environment, but might be hired for jobs related to the 3Ds (difficult, dangerous, and dirty) [3]. Migrant workers also commonly experience workplace injuries, long working hours, inadequate wage, salary denial, and restricted freedom [4]. It is also common to experience physical, sexual, and psychological violence by employers and brokers [5, 6]. All these maltreatments and working conditions expose migrants to poor health conditions, especially to poor mental health conditions. The situation would be worsened when migrants are repatriated in times of crisis such as the COVID-19 pandemic and when they pass through unexpected life challenges like compulsory quarantine [7, 8]. The discrepancy between the initial expectations of migrants and what they actually experienced could also exacerbate their mental health condition [9].

Studies showed that Ethiopian migrant returnees are victims of the negative consequences of irregular migration indicated above [10–12]. Evidence also revealed that because of public expectations about the stressful and unfriendly conditions that migrants experience while they were abroad, migrant returnees are usually perceived and treated as mentally incompetent in their own community [13]. Consequently, migrant returnees usually experience public stigma and discrimination and could also suffer from self-stigma that ultimately worsens their psychological distress unless appropriate coping strategies are employed.

There are some studies that showed the magnitude of and factors associated with psychological distress [11, 14] and coping strategies among migrant returnees in Ethiopia [12]. However, evidence is scarce about the magnitude of stigma and discrimination and associated factors, particularly among Ethiopian migrant returnees. Available studies looked at factors associated with psychological distress and coping strategies separately, not concomitantly. We found one study that documented the concomitant association among depression, anxiety, and post-traumatic stress disorder versus socio-economic and trafficking exposure characteristics; however, the study did not incorporate migration-related stigma and discrimination and coping strategies in the analysis [10] which justifies the need to examine all these variables concurrently. Hence, the present study intends to investigate the mediating role of psychological distress in the association between migration experiences and stigma and discrimination and coping strategies.

The need to see relationships among different variables simultaneously is necessary since different phenomena such as health and social problems are not independent of each other in the real world situation but rather interlinked, and thus relationships are not simple. As a result, examining the real complex network of relationships among these psychosocial problems with other potential factors concomitantly is justified. Therefore, the current study aimed to determine the mediating role of psychological distress in the relationship between migration experiences and stigma and discrimination and coping strategies using structural equation modeling. The study also aimed to find out the magnitude of psychological distress, stigma and discrimination and coping strategies among returnees.

Methods

Study setting and period

The study was conducted in five migration hotspot areas in Ethiopia, and data were collected between April and May, 2022. The study sites included Addis Ababa, Dessie, Shashemene, Hossana, and Gondar representing the different regions of the country. Participants recruited from Addis Ababa were expected to represent migrant returnees living in Addis Ababa and the central part of the country. Similarly, returnees recruited from Dessie city are expected to represent the northeastern part of the country, and returnees sampled from Shashemene and Hossaena represent the southeastern and southern part of the country, respectively. Returnees recruited from Gondar city represent returnees in the Northwestern part of the country. Participants were considered eligible to the study if they come from either the Middle East /Arab countries or South Africa between December 2019 (after the outbreak of COVID-19) and September 2021. In addition, they need to be residents of the study sites mentioned above during the data collection period. However, their place of departure at the initiation of migration might be anywhere in the country.

Study design

The current study is a cross-sectional survey that examined the mediating role of psychological distress in the relationships among migration experiences, coping strategies, and stigma and discrimination among returnees who came back to their home country during the COVID-19 pandemic period. Participants were approached through relevant government and NGO offices working on the reintegration of migrant returnees in the five sites. Returnees were asked to recall any psychological health issues they had as well as any coping strategies they practiced. They were also requested to indicate if they had experienced any stigma and discrimination just because they were migrant returnees in addition to other potential factors such as on socio-economic and migration experiences related variables.

Participants and sampling

The study population included Ethiopians who migrated to the Middle East and South Africa looking for opportunities related to their livelihood concerns, and who returned home between December 2019 and September 2021. For each study participants, a minimum of six months period elapsed between the dates of return and interview so that there would be sufficient period of time to show any experience of psychological distress, stigma and discrimination, and utilized of coping strategies after their return. To meet this requirement, the last possible dates of return and the initiation of data collection period were set as September 2021 and April 2022, respectively.

To guide the identification of variables, sample size determination, analyses, and reporting of findings, a hypothesized model was developed. The model represents the concomitant relationships among psychological distress, coping strategies, and migration-related stigma and discrimination as well as the mediating role of psychological distress in the relationship of sociodemographic characteristics and migration experiences with stigma and discrimination and coping strategies. It was developed after referring to literature [8, 10, 11] and considering the temporal relationships among variables. Psychological distress referred to the psychological status of participants after return home country. Whereas coping strategies would be practiced by returnees after the problem (psychological distress) was developed.

The study used a sample of 739 returnees which was determined for a larger research project related to migration experiences. Authors confirmed that this sample size is also sufficient to address research questions of the current study that used a structural equation modeling (SEM) to analyze it concomitantly. For SEM, the rule of thumb to calculate the minimum but adequate sample size is the N:q ratio rule (58), where N is the number of participants and q is the number of model parameters that require statistical estimates which ranges from 5—20 times the number of free parameters to be estimated in a hypothesized model [15]. Considering our hypothesized model (Fig. 1), there would be 48 parameters to be estimated (4 variance of independent variables, 6 factor loadings, 9 regression coefficients, 18 error variance, 8 covariance of independent variables, 3 error variances of latent variable). Taking N: q ratio of 10, the required sample size would be 480. Because we used two stage sampling, the first being sample study sites and the second sample returnees, the authors assumed a design effect of 1.5 which gave us the final sample size of 720. So, our sample size of 739 is close to the theoretically sufficient sample size showing that it is adequate.

Fig. 1.

A hypothesized model for the association of variables including psychological distress, discrimination, and coping strategies

The study sites (i.e. Addis Ababa, Dessie, Shashemene, Hosaena, and Gondar) were selected purposively taking into account their higher prevalence of transnational migration [16, 17]. Once the study sites were determined, the respective local administrators were approached to provide the research team with the list of migrants who returned home between December 2019 and September 2021 together with their contact addresses. Thus, sample returnees were recruited consecutively until about 150 were interviewed in each study site.

Variables and measurement

The independent or exogenous variables that were analyzed in the current study are migration experience related variables, namely physical violence experienced at destination, restricted freedom of movement, denial of salary as a reason to return home, and destination country. Among the outcome variables, psychological distress was the mediator variable and coping strategies and stigma and discrimination were the endogenous variables.

Psychological distress was measured with the 21-item version depression, anxiety and stress scale (DASS-21). The measure has three sub-scales (i.e. emotional states of depression, anxiety and stress) and each subscale has seven items. Each item in DASS-21 has four possible responses: ‘did not apply to me at all’ (assigned a score of 0), ‘applied to me to some degree (score of 1)’, ‘applied to me to a considerable degree’ (score of 2), and ‘applied to me most of the time’ (score of 3). The global Cronbach alpha of the DASS-21 instrument was estimated at 0.94, and 0.85, 0.89 and 0.89 for the depression, anxiety, and stress sub-scales, respectively.

Stigma and discrimination was measured with a 13-item instrument assembled from two standardized and validated stigma and discrimination measures originally developed for HIV/AIDS and leprosy patients. The instrument has three sub-scales: internalized stigma (five items), anticipated stigma (three items), and experienced stigma (five items). The internalized stigma subscale measures the degree to which participants endorse prejudicial beliefs and stereotypes [18] about people living with a certain condition. Participants were asked how frequently they feel the emotions indicated in the five items, on a scale from 0 (never) to 3 (often). The items were adapted to make them suitable for the migrant population. Three of the internalized stigma items, for example, were “I feel shy and ashamed of my being a migrant returnee”, “I feel like I am unwanted person”, and “I feel rejected by others because of my migration experience.” The reliability of the overall stigma and discrimination scale was confirmed with Cronbach alpha and estimated at 0.93. Similarly, the internalized stigma sub-scale had a Cronbach alpha of 0.90, anticipated stigma 0.71, and experienced stigma 0.87.

Anticipated stigma is expecting to experience prejudice, discrimination, and stereotyping from others in the future [19, 20]. It was psychometrically validated on patients in Ethiopia, and was found to be valid and reliable [21]. The internalized, anticipated, and experienced stigma scores were derived from the respective number of items to determine the average scores of each subscale.

The coping strategies of migrant returnees with a stressful life event was measured with a global 28-item questionnaire (coping orientation to problems experienced inventory, Brief-COPE) with three subscales, namely problem-focused coping (8 items), emotion-focused coping (12 items), and avoidant coping (8 items). All items in each subscale were scored on a 4-point Likert scale ranging from 1 (I haven't been doing this at all) to 4 (I’ve been doing this a lot). Both a mean and a total score were calculated for each subscale and the global scale [20]. The global Cronbach alpha of Brief COPE instrument was estimated at 0.76. The Cronbach alpha of problem focused sub-scale was 0.79, emotion focused was 0.76, and avoidant coping was 0.66. To better understand the coping strategies that the participants are practicing, the scores of 14 facets that contain two items each were summarized and measured with average (SD) [22]. A facet with an average close to 4 indicates that it was predominantly practiced to cope with the challenge that participants faced. Generally, all the tools used in this research work were internally consistent though were not validated on migrant returnees in Ethiopia. Migration experience related variables, including physical violence experienced at destination, denial of salary as a reason for return, and restricted freedom were measured using a questionnaire developed and tested by authors of this manuscript and were categorized as ‘not experienced at all’, experienced somehow’, and ‘experienced very much’.

Data collection procedure

All of the measures used in the study were originally compiled in English and then translated into Amharic. Both the original and translated versions of the questionnaire were reviewed by the study team and amendments were made. We did a pilot study in Addis Ababa using face-to-face interviews in order to determine the feasibility of the study and the adequacy of the questionnaire. This small-scale assessment was carried out with 15 migrant returnees who were eligible for the study as per the inclusion criteria identified earlier. Before the start of the main survey, the questionnaire was revised in accordance with the results of the pilot study.

Four data collectors and two supervisors were assigned in each of the five study sites during the main survey. The data collectors were from both sexes and were at least bachelor degree holders. The supervision was carried out by members of the research team who were conducting the larger thematic research project from which the current quantitative component is drawn. A one-day long training was given to the field workers focusing on the questionnaire and procedure of the interview. During data collection, convenient interview sites were chosen by the data collectors to encourage participants express themselves and their experiences freely. Interpreters were used in cases where data collectors did not speak the local language. The data collection was strictly overseen by the investigators and field supervisors.

Data processing and analysis

The completed questionnaires were manually reviewed for accuracy. Data were entered into and coded and cleaned with Statistical Package for Social Science (SPSS) and exported to Stata version 14 for further analyses. Descriptive statistics (mean, standard deviation (SD), number (percent), and minimum–maximum values) were used to summarize the findings as appropriate.

The global and sub-scales internal consistency reliability of DASS-21, stigma and discrimination scale, and the Brief-COPE were assessed using Cronbach’s alpha. Internal consistency reliability is considered satisfactory if Cronbach’s alpha is > 0.7. To measure association between variables, unstandardized coefficients, 95% confidence level, and 5% level of significance were used. For selected associations, standardized beta-coefficients were estimated and discussed for assessing the relative influence on the response. For each sub-scale, the scores of the items were used without any transformation to estimate Cronbach’s alpha. However, for SEM analysis parceling was performed by creating a composite variable using items in each subscale [23].

After parceling, the three (global) scales were still latent variables each constituting the respective subscales summarized with composite variables as their indicators or items. Since variables with four or more levels can be considered as continuous, the measurement models were analyzed considering the items as Gaussian variables [24, 25]. Thus, SEM was employed to examine the relationship among various exogenous and endogenous or mediating variables. Diagrammatically, the correlation among disturbances (residual errors that reflect the unexplained variances in the latent endogenous variables due to all unmeasured causes) was indicated by double arrows, and the association of each exogenous or mediating variable with the corresponding dependent variable was shown by the path coefficient along with a single headed arrow.

When direct and indirect effects were present due to mediation of effects, the total effects were determined using the linear combination of estimator technique. The SEM used maximum likelihood with missing values estimation method. The analysis was started with the hypothesized model (Fig. 1), and modifications were performed iteratively by removing or adding path links or including mediator variables if theoretically supported until the final model (Fig. 2) that adequately fitted the data and seemed conceptually meaningful was retained.

Fig. 2.

The mediating of psychological distress between migration experience and stigma and discrimination and coping strategies

Results

Background characteristics of participants

A total of 739 migrant returnees were included in the study with a response rate of 100%. The majority of the participants (70.50%) were females. The average age of the participants at the time of departure was 22.21(SD = 4.84), but at the time of interview was 30.56 (SD = 6.34) with minimum and maximum of 19 and 60 years, respectively. About 341 (46.39%) returnees were Muslim. While 480 (65.31%) were never married at the time of departure, 406 (55.09%) were married at the time of data collection. Only 301 (41.35%) completed primary education, 419 (57.95%) were urban residents prior to departure, about one-third or 247 (33.56%) of the participants migrated to Saudi Arabia, and the average length of stay abroad was 6.38 years (SD = 4.31) (Table 1).

Table 1.

Background characteristics of participants

Characteristics	Number (%)
Age at departure (n = 737); Mean = 22.21 and SD = 4.84
10–17	92 (12.48)
18–25	497 (67.44)
26–30	111 (15.06)
31–50	37 (5.02)
Current age (n = 737); Mean = 30.56 and SD = 6.34
19–25	148 (20.08)
26–30	294 (39.89)
31–40	245 (33.24)
41–60	50 (6.78)
Sex (n = 738)
Male	217 (29.40)
Female	521 (70.60)
Religion (n = 735)
Muslim	341 (46.39)
Orthodox	226 (30.75)
Protestant	155 (21.09)
Other	13 (1.77)
Marital status at the time of departure (n = 735)
Married	211 (28.71)
Never married	480 (65.31)
Separated/Divorced/Widowed	44 (5.98)
Current marital status (n = 737)
Married	406 (55.09)
Never married	240 (32.56)
Separated/Divorced/Widowed	91.1 (12.35)
Educational level (n = 728)
Cannot read and write	73 (10.03)
Can read and write but no formal education	35 (4.81)
Primary	301 (41.35)
Secondary	280 (38.46)
College or university	39 (5.36)
Region of current residence (n = 739)
Addis Ababa	111 (15.02)
Amhara	319 (43.17)
Oromia	125 (16.91)
SNNP	147 (19.89
Other	6 (0.82)
Residence before departure (n = 723)
Rural	304 (42.05)
Urban	419 (57.95)
Parents alive at the time of departure (n = 735)
Both parents alive	509 (69.25)
Only father alive	50 (6.80)
Only mother alive	129 (17.55)
Both were not alive	47 ( 6.39)
Occupation before departure (n = 738)
No job at all	183 (24.80)
Student	343 (46.48)
Daily laborer	53 (7.18)
Employed (private/government)	43 (5.82)
Self-employed or doing own or family’s job	82 (11.11)
Other	34 (4.61)
Parents’ means of income before departure (n = 738)
No parents or guardian	11 (1.49)
Government employee	40 (5.42)
Private firm employee	35 (4.74)
Farming	462 (62.60)
Trade	112 (15.18)
Other	78 (10.57)
Destination country (n = 736)
Saudi Arabia	247 (33.56)
Lebanon/Beirut	128 (17.39)
Qatar	23 (3.13)
Kuwait	65 (8.83)
South Africa	102 (13.86)
Sudan	67 (9.10)
Dubai/United Arab Emirates	73 (9.92)
Other	31 (4.21)

Psychosocial problems and coping strategies

Among psychological problems, the average (SD) score of depression was 1.20 (0.83), anxiety 1.03 (0.85), and stress was 1.27 (0.80). Concerning coping strategies, the average (SD) of problem-focused coping was estimated at 2.73 (0.62), emotion-focused coping 2.53 (0.53), and avoidant coping was 2.14 (0.54). To pinpoint in-depth the styles of coping that the returnees were practicing, the scores of 14 facets that contain two items under each were summarized. Accordingly, the average (SD) of coping with religious practice was 3.54(0.67), planning was 2.90(0.79), and positive framing was 2.85(0.81) (Table 2).

Table 2.

Psychological problems and facets of coping strategies of returnees

Characteristics	Mean (SD) of scores
Psychological problem
Depression (n = 689)	1.20 (0.83)
Anxiety (n = 715)	1.03 (0.85)
Stress (n = 699)	1.27 (0.80)
Internalized Stigma (n = 720)	2.04 (0.95)
Anticipated Stigma (n = 713)	2.31 (0.89)
Experienced Stigma/discrimination (n = 704)	2.17 (0.94)
Facets of styles of coping
Religion (n = 731)	3.54(0.67)
Planning (n = 719)	2.90(0.79)
Positive reframing (n = 731)	2.85(0.81)
Self-distraction (n = 719)	2.82(0.82)
Active coping (n = 728)	2.79(0.79)
Acceptance (n = 726)	2.57 (0.87)
Venting (n = 723)	2.46(0.91)
Self-blame (n = 726)	2.43(1.01)
Use of informational support (n = 717)	2.38(0.95)
Denial (n = 719)	2.29(0.88)
Emotional support (n = 729)	2.13(0.86)
Humor (n = 721)	2.01(0.99)
Behavioral disengagement (n = 725)	1.98(0.88)
Substance use (n = 731)	1.47(0.84)

Associations among background characteristics, migration experiences, psychological distress, stigma and discrimination and coping strategies

After a series of specifications and refitting of various theoretically meaningful candidate models iteratively, the final model showing the relationship among the three latent variables, namely psychological distress, stigma and discrimination, and coping strategies of returnees in relation to migration experience related variables was retained (Fig. 2). The overall model fitness test showed that the root mean square error of approximation (RMSEA) = 0.064 and the comparative fit index (CFI) was = 0.965.

The final model included four exogenous variables (physical violence at the destination, denial of salary by employers as a reason to return home, restricted freedom, and destination country), one mediator variable (psychological distress), and two endogenous variables (stigma and discrimination and coping strategies). Denial of salary by employers as a reason to return and restricted freedom were both directly and indirectly related through the mediating variable (i.e. psychological distress) with stigma and discrimination. Similarly, being a returnee from South Africa was both directly and indirectly related through psychological distress with coping strategy.

The multivariate analysis confirmed that physical violence experienced at destination had a direct positive association with psychological distress (adjusted β = 1.99, 95% CI: 1.20, 2.78). Returnees who migrated to South Africa had psychological distress score less than that of returnees from the Middle East, i.e. less by a score of more than two (adjusted β = -2.11, 95% CI: -3.20,—1.03). Denial of salary by employers had a positive direct (adjusted β = 1.00, 95% CI: 0.50, 1.50), indirect (adjusted β = 1.20, 95% CI: 0.71, 1.68), and total (adjusted β = 2.20, 95% CI: 1.54, 2.85) association with stigma and discrimination. Denial of salary had also a positive association with psychological distress (adjusted β = 1.99, 95%CI: 1.20, 2.78 or adjusted and standardized β = 0.19, 95%CI: 0.01, 0.28.

In terms of destination, returning from South Africa had a direct (adjusted β = 2.50, 95% CI: 1.66, 3.33), indirect (adjusted β =—0.61, 95% CI: -0.95, -0.28), and total (adjusted β = 1.89, 95% CI: 1.02, 2.75) association with coping strategies. Similarly, restricted freedom experienced at the destination had positive direct (adjusted β = 0.54, 95% CI: 0.28, 0.81) or adjusted and standardized β = 0.12, 95%CI: 0.04, 0.20), indirect (adjusted β = 0.39, 95% CI: 0.13, 0.65) and total (adjusted β = 0.94, 95% CI: 0.59, 1.29) association with stigma and discrimination.

Psychological distress was positively and significantly associated with stigma and discrimination (adjusted β = 0.60, 95%CI: 0.54, 0.66) or had adjusted and standardized β = 0.72 (95%CI: 0.63, 1.62). It had also association with coping mechanisms with adjusted β = 0.29 (95%CI: 0.23, 0.35) or adjusted and standardized β = 0.39 (95%CI: 0.34, 0.44).

Finally, returnees from South Africa have lower risk of developing psychological distress when compared to those from the Middle East and other Arab countries (adjusted β = -2.11, 95%CI: -3.20, 1.03 or adjusted and standardized β = -0.14, 95%CI: -0.23, -0.06); they are also more likely to practice adaptive coping strategies more frequently than their counterparts to overcome their psychological distress.

The covariance between the residual errors of anticipated and experienced stigma (covariance = 1.39, 95% CI: 0.87, 1.91) as well as that of problem focused and avoidant coping (covariance = -5.4, 95% CI: -8.01, -2.92) were statistically significantly (Table 3).

Table 3.

Direct, indirect, and total effects of migration experience variables on psychosocial problems and coping mechanisms among Ethiopian migrant returnees (n = 628)

Characteristics		Direct Effect β (95% CI)	Indirect Effect β (95% CI)	Total effect β (95% CI)
Dependent variable: Psychological distress
Physical violence	1.02 (0.57, 1.47)		-	1.02 (0.57, 1.47)
Destination
Middle East		0.0		0.00
South Africa		-2.12(-3.20, -1.03)	-	-2.12(-3.20, -1.03)
Restricted freedom		0.65 (0.23, 1.08)		0.65 (0.23, 1.08)
Denial of salary		1.99 (1.20, 2.77)	-	1.99 (1.20, 2.77)
Dependent variable: Stigma and discrimination
Psychological distress		0.60 (0.54, 0.66)		0.60 (0.54, 0.66)
Restricted freedom		0.54 (0.28, 0.81)	0.39 (0.13, 0.65)	0.94 (0.59, 1.29)
Denial of salary		1.00 (0.50, 1.50)	1.20 (0.71, 1.68)	2.20 (1.54, 2.84)
Destination
Middle East			0.0	0.00
South Africa			-1.27 (-1.93, -0.61)	-1.27 (-1.93, -0.61)
Physical Violence			0.61 (0.34, 0.89)	0.61 (0.34, 0.89)
Dependent variable: coping strategy
Psychological distress		0.29 (0.23, 0.35)		0.29 (0.23, 0.35)
Destination
Middle East and Arab		0.0		0.00
South Africa		2. 50 (1.66, 3.33)	-0.61 (-0.95, -0.28)	1.89 (1.02, 2.75)
Physical violence			0.29 (0.15, 0.44)	0.29 (0.15, 0.44)
Restricted freedom			0.19 (0.06, 0.32)	0.19 (0.06, 0.32)
Denial of salary			0.57 (0.32, 0.83)	0.57 (0.32, 0.83)

CI confidence interval, β beta-coefficient (unstandardized)

Discussion

The current study shows that a considerable number of migrant returnees are affected by psychological distress and stigma and discrimination. However, lower than expected number of returnees are practicing favorable coping strategies to get out of the problem. Religious coping strategy, a variety of emotion focused coping, was the most commonly used coping mechanism followed by planning which is a problem focused coping. We found that psychological distress significantly associated with both stigma and discrimination and coping strategies. More importantly, psychological distress is found to be a significant mediator variable in the relationship between migration experiences, namely physical violence, denial of salary by employers as a reason for return, restricted freedom, and destination country and stigma and discrimination and coping strategies. Thus, psychological distress shapes the association between migration experiences and stigma and discrimination and coping styles.

In this study, we found that over half of migrant returnees reported symptoms of depression (57.33%) and anxiety (59.86%); however, less than one-half of them reported symptoms of stress (45.21%). The prevalence of reported depression and anxiety symptoms in this study is much higher than that of prevalence estimates in the general population in Ethiopia (i.e. 4.7% and 3.3%, respectively) [26]. The prevalence estimates in the current study are, however, comparable to our other research among migrant returnees who were under mandatory quarantine (depression was 55% and anxiety was 48.9%) [11] but slightly less than that of another study conducted among trafficked migrant returnees in which 58.3% of them were reported to have depression and 51.9% anxiety [10]. Of course, it is not surprising to have a high magnitude of psychological distress when compared to the general population of Ethiopia because migrant returnees have been exposed to tragic experiences during their migration period that could result in mental disorders unlike that of the general population who mostly are leading their normal life.

The very high magnitude of psychological distress among participants of the present study may also be associated with the timing of their return and the overall situation in their home country. The participants repatriated to their country during the time of the COVID-19 pandemic, in which some of them were required to stay in a fourteen-day mandatory quarantine, some were refused by parents for reunion due to fear of the virus, and some were unable to reunite with their parents due to the ongoing conflict in the country. On the other hand, it may be justifiable to have lower estimates of psychological distress (in particular the anxiety estimate) than trafficked returnees as trafficked persons might have been exposed to extra burden and mistreatment than the common irregular migrants due to the experience of exploitation by traffickers [27, 28].

The comparable estimate of depression and anxiety between the current study and that of our previous research on participants who were under mandatory quarantine may seem unexpected because study participants in the latter investigation were under compulsory quarantine during the COVID-19 outbreak. Even the magnitude of stress in the current study (59.66%) was much higher than the previous one which was reported as 35.6% [11]. Probably, this may be because returnees in the current study stay longer at home which may be less comfortable to them due to the higher stigma and discrimination that they have reportedly experienced and the limited availability of occupational and social opportunities.

The mean of emotion focused coping strategy was 2.53, and specifically religious coping (average = 3.54) was the most commonly used strategy followed by planning (average = 2.90) which is a problem focused coping. The mean score of problem focused coping strategies was estimated at 2.73 which indicates a medium level of utilizing coping strategies that were aimed at changing the stressful situation. This implies that the returnees’ psychological strength and practical approaches they used to cope with psychological problems are not sufficiently high, and thus, it is less likely to experience positives outcomes. On the other hand, there was a low score (i.e. 2.14) of avoidant coping strategies. This implies the little physical or cognitive efforts that the returnees made to disengage themselves from the stressors. Such low scores are typically suggestive to adaptive coping [22].

Psychological problems and subsequent social problems such as stigma and discrimination are complex [29] and the current study addressed this situation by examining the interwoven relationships concomitantly using SEM. Accordingly, three contextual variables, namely psychological distress, stigma and discrimination, and coping strategies, each with three subdomains, were analyzed using SEM [28] considering other potential predisposing factors that were mainly migration experiences [27, 28, 30, 31]. Thus, this study contributed considerably to the existing body of knowledge by employing the appropriate statistical methods to explain the complex network of relationships among all these variables.

Psychological distress, which constitutes anxiety, depression, and stress, is found to be positively and significantly associated with stigma and discrimination (adjusted β = 0.60, 95%CI: 0.54, 0.66) and coping mechanisms (adjusted β = 0.29, 95%CI: 0.23, 0.35). It is also found to be a mediator variable in the relationship between migration experiences and stigma and discrimination and coping mechanisms. Because they are just migrant returnees, local people in the home country attach with them a stereotype that considers them mentally incompetent [13]. Because of this stigma, returnees could be discriminated from various opportunities and services that would in turn exacerbate their preexisting mental health problems or could develop mental health symptoms as a result of the mistreatments. This implies that psychological distress after return might be the result of or at least be intensified by stigma and discrimination by experiences in the local community or vice versa.

The positive association between psychological distress and coping strategies could be justifiable in that one can device a coping strategy if there is only a problem which is in this case a stereotype associated with merely their being migrant returnees. If the level of the distress is extremely severe, however, it might be difficult for a person to respond with a coping strategy and pursue it properly for better outcome.

The current study revealed that only migration-related exposures that were experienced at the destination country, but not those experienced during travelling stages, were found to be associated with psychological distress, stigma and discrimination, and coping strategies. Similarly, background or pre-departure characteristics do not have significant association with these psycho-social and coping strategy characteristics. In this context, physical violence experienced at the destination country is positively associated with psychological distress. The association could probably be related to the extra force used by employers or the mistreatment returnees experienced at the destination stage which usually results in psychological problems [27].

Denial of salary by employers and restricted freedom in the destination country are shown to be associated with both psychological distress and stigma and discrimination, and psychological distress is found to be a mediator variable for each of these associations. The statistically significant relation between restricted freedom and psychological distress (adjusted β = 0.65, 95%CI: 0.23, 1.08) could probably be linked to the stressful conditions such as strict control exerted over migrants that limit their right to behave and move freely [27, 28]. It may be justifiable for the positive association between the denial of salary and psychological distress (adjusted β = 1.99, 95%CI: 1.20, 2.77) in the current study as it may upset or turn down the moral and interest of returnees.

Finally, returnees from South Africa have lower risk of developing psychological distress when compared to those from the Middle East and other Arab countries (adjusted β = -2.12, 95%CI: -3.20, -1.03); they are also more likely to practice adaptive coping strategies more frequently than their counterparts to overcome their psychological distress. This may be related to context differences between the two destinations, the Middle East and South Africa, where the former might have exposed migrants to various inconveniences leading into psychological distress. In addition, it is well documented that returnees from South Africa are mostly men [9, 17] while the participants in this study are mostly women. Women are generally reported to have a higher risk of developing mental distress compared to men [32, 33].

Strengths and limitations

This study empirically and concomitantly tested the mediation of psychological distress in the relationship between migration experiences and stigma and discrimination and coping strategies with a large sample in Ethiopia. However, the authors want to warn the readers that cross-sectional approaches to mediation typically generate biased estimates [34, 35]. Some of the items in the DASS-21, stigma and discrimination, and Brief-COPE scales are sensitive, and hence migrant returnees are unlikely to endorse them and as a result there could be under reporting of psychological distress. However, we have made efforts to inform respondents about the confidentiality of personal information during the interview to encourage them in giving accurate information. In addition, as we asked participants about their past experiences retrospectively, there might be recall bias though we understood that most of their experiences were linked with marked events in their life. The other limitation of the study was the use of a screening tool to measure psychological distress but not a diagnostic tool. A further limitation is the use of a mental health screening tool and other instruments which have not been validated among migrant returnees, at least in the context of Ethiopia. However, the validity and reliability of the tool was investigated in the general population and in culturally diversified contexts. Besides, the tools were used in different researches including migrant returnees and were internally consistent [8, 11]. We also found that the reliability of all of the scales and sub-scales we used in this study are in acceptable levels.

Implications

The current study shows that psychological distress is important not only as an outcome or predictor variable among migrant returnees; it has also a mediating role in the association between migration experiences and stigma and discrimination and coping. This implies that migrant returnees are experiencing stigma and discrimination not just because they are returnees but may be because they have depression and anxiety symptoms. Migration experiences such as violence in destination, denial of salary and restricted freedom are likely to increase mental health symptoms and these again bring about stigma and discrimination. The study also shows that psychological distress shapes the coping strategy that migrant returnees are likely to use.

Overall, the findings of this study suggest that migrant returnees need to access mental health services to deal with their psychological distress and then to reducing their stigma and discrimination and to enhance their coping skills. Mental health services need to be part of the reintegration efforts of stakeholders for migrant returnees. However, further research is needed to identify contextually suitable psychosocial interventions to treat psychological distress among migrant returnees in Ethiopia.

Conclusions

We found that the magnitude of psychological distress among Ethiopian migrant returnees is high. Internalized, anticipated, and experienced stigma are considerably common in this population. Religious practice was the most commonly used coping mechanism. The returnees’ psychological strength and practical approaches to cope with psychological problems and the associated stigma and discrimination are not sufficient.

Psychological distress has significant positive association with both sigma and discrimination and coping strategies. Physical violence experienced at the destination, denial of salary by employers as reason to return, and restricted freedom were positively associated with psychological distress which is also a mediating factor in the relationship between migration experiences and stigma and discrimination and coping strategies. Returnees from South Africa managed to reduce psychological distress and practice positive coping strategies more than those who returned from the Middle East. Thus, returnees from the Middle East may need special attention to their psychological problem, and generally returnees need psychosocial interventions to reduce their stigma and discrimination, and to enhance their utilization of useful coping strategies. Besides, we recommend the validation of the tools and the respective cut-off points of scores used in Ethiopian context especially on migrant returnees.

Abbreviations

DASS-21

21 Item version Depression, Anxiety and Stress Scale

Brief-COPE

Brief Coping Orientation to Problems Experienced

3Ds

Difficult, Dangerous, and Dirty

COVID-19

Coronavirus Disease of 2019

NGO

Non-Governmental Organization

SEM

Structural Equation Modeling

SD

Standard Deviation

SPSS

Statistical Package for Social Science

CI

Confidence Interval